Two Missing Considerations in the Inequality Debate

Recently I have been engaging in a lot of "bashing the inequality crowd"-type posts (see my posts on Wahlroos, Oxfam or measurement errors) at the expense of my many other intellectual interests. This is justified, I'd argue, since Life of an Econ Student and the experience of being an econ student in the 21st century is to constantly argue or consider inequality – even in class! It is a bit like historians and their relationship to class-gender-race. All. The. Bloody. Time. 

Now, there are many shortcommings to the inequality discussion in general: the lack of a global perspective (Milanovic), the neglect of human capital (McCloskey, Goldin & Katz), the constant jumping between income, wealth and consumption inequalities to form an argument (Horwitz), the inconclusive data problem in general (Snowdon), how little it matters for real people IRL, or the inconsistent double-standard treatment of the subject made famous by Krugman and attacked by Cochrane. Here's the main point of my ranting post 'Off With Their Heads' in November:

If most of the harm of inequality, Pickett-Wilkinson style, is transmitted through a social envy process, where my poor health outcomes are blamed on the fortunes of the Bill Gates and Warren Buffets of the world (or even my property-managing next-door neighbour), it is crucial that I see them. If, as Krugman's article points to, most people are so greatly unaware of the extent to which incomes are unequally distributed, they cannot be affected by them in that way; if the rich are so removed from my ordinary life, Dr. Krugman, then the envy-keep-up-with-the-Joneses process that increases my stress and worsen my mental health cannot occur. If I suffer from seeing wealthy people, but I hardly ever see wealthy people, then conclusively, I do not suffer very much.

I often wonder why we're still talking about inequality, but there are many things in 21th century academia and political life that I still have to grasp, so #patience is a thing. Anyway, I recently stumbled upon two points to the inequality debate that are hardly ever (read: never) considered. More importantly, if one would consider them, the entire inequality disaster story falls apart.

First, Harvest economist Martin Feldstein argues that future pension payments are not included into the consideration of wealth or even total lifetime income (non-paywalled version here or here). Beyond the interesting discussion of tax code changes and their impact on measured inequality, his point is strikingly obvious, yet often unconsidered (Brad DeLong makes the clueless objection that Feldstein's critique is good, but still "everybody knows income inequality is rising, so doesn't matter" – excuse me?): most workers, especially those at the lower end, have future pension & social security claims on the government. While some of those debts are likely to be defaulted on, at least some share of them should be attributed to workers/low-income earners' lifetime income. What Piketty (and Oxfam...) and others end up doing by not including them is not simply comparing apples and pears  – it's comparing apples and, like, combustion engines! Feldstein's take-away point:

That wealth includes the present actuarial value of Social Security and retiree health benefits, and the income that will flow from employer-provided pensions. If this wealth were taken into account, the measured concentration of wealth would be much less than Mr. Piketty's numbers imply.

By benefit caps, progressive taxation and the way these benefit systems are set up, they are progressive in nature; as a share of total life-time income, pensions and social security will constitute a much bigger part for low-income earners than for high-income earners, which means that incorporating Feldstein's critique diminishes (or perhaps even reverses, if the magnitudes are large enough) the Piketty arguments of income distribution. In either case, this is another way in which the 'inequality crisis' is exaggerated.

Second, as my brilliant flatmate pointed out to me during one of our late-night arguments, flushed out further on his own blog:

As easy as it is to observe capital-related wealth by using share or bond price indices, labour-related wealth remains hidden when we account for wealth distributions using classical models. No wonder that wealth numbers indicate a more unequal, highly dispersed scale compared to plain vanilla income figures.

In the case of a financial claim the market simply calculates a net-present value of it, and right away we know the immediate wealth of the individual(s). But what does this really mean? According to Benjamin Graham’s famous metaphor Mr. Market tells us a price where he is willing to buy our future cash-flow. There is no difference in the case of labour-related income either. The market or bank gives us a quote on which we can exchange our future earnings for a certain amount of money. Yet this kind of wealth is not accounted for, a missing element of household balance sheets which distorts overall numbers when compared to capital instruments.

Let me unpack the conceptual brilliance here: wealth (as held by the Bill Gates or Warren Buffets of the word) are mostly held in stocks or bonds. The value of a stock depends on the underlying company's ability to earn profits, with appropriate financial market risk taken into account. Similarly, the value of a government bond depends on the ability of the underlying government to levy taxes on its economy and service its debt. That is, the value of the financial instrument reflects the future earnings potential of the company or state considered.

In wealth inequality figures, we compare those fortunes of the rich to... absolutely nothing. For labour, we have no such market-traded asset for which we could compare fortunes. A similar treatment for the "labour stock" to how we treat capital, would be to value the future earnings potential of labour, pack it into an instrument and value it accordingly. Because this is exclusively owned by the workers – and not traded publicly  – our comparison is between something we can observe (wealth, capital) and something we cannot observe (labour). When he did the math, deriving the hypothetical value of U.S. future labour (using Gordon growth model), my flatmate found that it would be in the ballpark of a million dollars for every living American:

So let me put it into context: this is 19-folds the GDP, or 14 times the S&P market capitalisation (all at the end of 2015). 

(Interestingly enough, the growth rate of this "labour stock" has outpaced S&P 500 returns over the last 50 years, and so Piketty's infamous r > g suffers another blow.) Labour stock would, in other words, completely swamp wealth and assets as currently measured. 

But aren't these just as unequally distributed as income? Wouldn't incorporating this "labour stock" simply extend the inequality by the same magnitude, like, say, doubling everyone's income (an operation that would leave most inequality measures intact)?

No. This would only be true if labour income (=wage) represented the same fraction of income across the income distribution, i.e. if both a rich and poor person earns, say, 90% of total incomes from wages. This is not the case, since income from labour at the high and very high end is a much smaller share of total income than labour income at the low end of the distribution (capital income, SME, options and share compensation etc.). That is, attempting to quantify the "labour stock" and adding it back into calculations of wealth inequality would result in a less unequal distribution  – perhaps even turning it upside down. 

In summary, if we would add back the largely neglect aspects of a) public pensions/Social Security and b) the appropriate comparison of "labour stock" into the considerations of wealth inequality, wealth inequality would fall dramatically. Conceptually this is an easy and straight-forward corretion and we could come up with plausible estimates (see above). Empirically it is however probably impossible, considering that trading "labour stock" in a market place (= the only way for us to get proper market valuations for them) would probably amount to some form of slavery (though, there are interesting arguments here).

Nevertheless, it is obvious that beyond the many many problems with the inequality crowd's arguments, these two massive elements have gone largely unnoticed and would greatly diminish the argument should they be taken into consideration.